Sharp quadratic χ-binding for powers of bipartite graphs

For every integer r2r\ge 2, does there exist a constant cr>0c_r>0 and an infinite family of bipartite graphs BnB_n such that ω(Bnr)\omega(B_n^r)\to\infty and χ(Bnr)crω(Bnr)2\chi(B_n^r)\ge c_r\,\omega(B_n^r)^2 for all nn, where BnrB_n^r is the rrth graph power? Equivalently, is the known quadratic upper bound χ(Br)=O(ω(Br)2)\chi(B^r)=O(\omega(B^r)^2) asymptotically sharp for every r2r\ge 2?

References

Progress summary

Refreshed
Claimed solved

A new preprint claims that the best possible growth is quadratic for every power of a bipartite graph, including squares.

The problem asks whether the known quadratic upper bound for coloring powers of bipartite graphs is sharp for every r2r \ge 2, including r=2r=2.

Known results

  • A 2023 preprint records the general quadratic upper bound for squares of bipartite graphs.
  • It gives bipartite graphs BB with χ(B2)=Ω ⁣(ω(B2)2logω(B2))\chi(B^2)=\Omega\!\left(\frac{\omega(B^2)^2}{\log \omega(B^2)}\right).
  • For convex bipartite graphs, it proves χ(G2)32ω(G2)\chi(G^2)\le \frac{3}{2}\omega(G^2).
  • It states that squares of biconvex bipartite graphs are perfect.

August 27, 2026 claimed sharpness

A new preprint, “Sharp quadratic χ\chi-binding functions for powers of bipartite graphs,” claims that the quadratic upper bound is best possible for every r2r\ge 2, closing the previous gap, including the square case. This claim is unverified.

Current status (as of August 2026): A new preprint claims the quadratic bound is sharp for every r2r \ge 2, but no independent verification is recorded.

Sources

Solutions 0

No solutions have been posted yet.